Image Denoising via Neighborhood Block Dimensionality Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital image processing methods for denoising images captured by CCD and CMOS image sensors, such as those using color filter arrays, face challenges in efficiently reducing noise while preserving image quality and computational efficiency, particularly in handling noise introduced by color filter mosaics.
Innovation Solution
The method employs neighborhood block dimensionality reduction to calculate denoised values for pixels by determining similarity between denoising pixels and reference pixels using characterization vectors, reducing computational complexity through selective averaging and non-selective averaging methods, and preserving edge details during the denoising process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If traditional denoising methods using neighborhood filters are applied to RAW image data, then noise reduction is achieved, but computational complexity increases significantly
Solution Approach 1:
The patent segments the neighborhood block into different regions (center pixel, surrounding pixels) and applies different processing strategies to each segment. The characterization vector is constructed by selectively averaging pixels from different segments, reducing the dimensionality from the full neighborhood block while preserving essential features for noise reduction.
Solution Approach 2:
The patent transforms the high-dimensional neighborhood block data into a lower-dimensional characterization vector by changing the parameter representation. Instead of using all pixel values, it uses averaged values from specific pixel groups, effectively reducing computational parameters while maintaining denoising capability.
2Measurement precision
If full neighborhood block data is used for denoising calculations, then denoising accuracy is improved, but processing time increases
Solution Approach 1:
The patent extracts only the essential features from the full neighborhood block by constructing a characterization vector. Instead of processing all neighborhood pixels, it extracts averaged values from strategically selected pixel groups, reducing processing time while maintaining the accuracy needed for effective denoising.
3Object-affected harmful factors
If color filter array information is used for denoising, then color-specific noise reduction is achieved, but device complexity increases
Solution Approach 1:
The patent applies local quality by treating different color channels (R, G, B) differently based on their specific noise characteristics. The characterization vector construction uses color filter information to selectively average pixels of the same color, applying localized processing strategies to each color component while maintaining overall system efficiency.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
System and method can support an image processing device. The image processing device operates to obtain a first set of characterization values, which represents a first group of pixels that are associated with a denoising pixel in an image. Also, the image processing device can obtain a second set of characterization values, which represents a second group of pixels that are associated with a denoising reference pixel. Furthermore, the image processing device operates to use the first set of characterization values and the second set of characterization values to determine a similarity between the denoising pixel and the denoising reference pixel. Then, the image processing device can calculate a denoised value for the denoising pixel based on the determined similarity between the denoising pixel and the denoising reference pixel.